Presents the fuzzy neural network optimized by hybrid coded genetic algorithm of decimal encoding and binary encoding, the searching ability and stability of genetic algorithms enhanced by using binary encoding during...Presents the fuzzy neural network optimized by hybrid coded genetic algorithm of decimal encoding and binary encoding, the searching ability and stability of genetic algorithms enhanced by using binary encoding during the crossover operation and decimal encoding during the mutation operation, and the way of accepting new individuals by probability adopted, by which a new individual is accepted and its parent is discarded when its fitness is higher than that of its parent, and a new individual is accepted by probability when its fitness is lower than that of its parent. And concludes with calculations made with an example that these improvements enhance the speed of genetic algorithms to optimize the fuzzy neural network controller.展开更多
Aiming at the problems that fuzzy neural network controller has heavy computation and lag,a T-S norm Fuzzy Neural Network Control based on hybrid learning algorithm was proposed.Immune genetic algorithm (IGA) was used...Aiming at the problems that fuzzy neural network controller has heavy computation and lag,a T-S norm Fuzzy Neural Network Control based on hybrid learning algorithm was proposed.Immune genetic algorithm (IGA) was used to optimize the parameters of membership functions (MFs) off line,and the neural network was used to adjust the parameters of MFs on line to enhance the response of the controller.Moreover,the latter network was used to adjust the fuzzy rules automatically to reduce the computation of the neural network and improve the robustness and adaptability of the controller,so that the controller can work well ever when the underwater vehicle works in hostile ocean environment.Finally,experiments were carried on " XX" mini autonomous underwater vehicle (min-AUV) in tank.The results showed that this controller has great improvement in response and overshoot,compared with the traditional controllers.展开更多
航空蓄电池作为飞机应急电源与启动系统的核心部件,其荷电状态(State of Charge, SOC)的精准估算对飞行安全与能源管理至关重要。针对传统单一算法(如安时积分法、扩展卡尔曼滤波(EKF)、神经网络法)在复杂工况下存在的模型依赖性强、非...航空蓄电池作为飞机应急电源与启动系统的核心部件,其荷电状态(State of Charge, SOC)的精准估算对飞行安全与能源管理至关重要。针对传统单一算法(如安时积分法、扩展卡尔曼滤波(EKF)、神经网络法)在复杂工况下存在的模型依赖性强、非线性拟合不足或累积误差大等问题,文章提出一种融合卡尔曼滤波与神经网络的复合算法。该算法以二阶RC等效电路模型为基础,通过卡尔曼滤波实现状态递推,并利用长短期记忆网络(LSTM)补偿模型参数漂移与非线性动态误差。实验基于某型航空镍镉电池实测数据(涵盖充放电循环、温度波动及老化场景)。结果表明,复合算法的SOC估算均方根误差(RMSE)降低至1.2%以内,较单一EKF与LSTM算法分别提升42%与35%,有效解决了航空场景下高动态、宽温域工况的SOC估算难题。展开更多
A novel approximation algorithm was proposed for the problem of finding the minimum total cost of all routes in Capacity Vehicle Routing Problem (CVRP). CVRP can be partitioned into three parts: the selection of vehic...A novel approximation algorithm was proposed for the problem of finding the minimum total cost of all routes in Capacity Vehicle Routing Problem (CVRP). CVRP can be partitioned into three parts: the selection of vehicles among the available vehicles, the initial routing of the selected fleet and the routing optimization. Fuzzy C-means (FCM) can group the customers with close Euclidean distance into the same vehicle according to the principle of similar feature partition. Transiently chaotic neural network (TCNN) combines local search and global search, possessing high search efficiency. It will solve the routes to near optimality. A simple tabu search (TS) procedure can improve the routes to more optimality. The computations on benchmark problems and comparisons with other results in literatures show that the proposed algorithm is a viable and effective approach for CVRP.展开更多
This paper presents an proportional integral (PI) based voltage-reactive power control for wind diesel based decentralized hybrid power system with wide range of disturbances to demonstrate the compensation effect on ...This paper presents an proportional integral (PI) based voltage-reactive power control for wind diesel based decentralized hybrid power system with wide range of disturbances to demonstrate the compensation effect on system with intelligent tuning methods such as genetic algorithm (GA), artificial neural network (ANN) and adaptive neuro fuzzy inference system (ANFIS). The effect of probabilistic load and/or input power pattern is introduced which is incorporated in MATLAB simulink model developed for the study of decentralized hybrid power system. Results show how tuning method becomes important with high percentage of probabilistic pattern in system. Testing of all tuning methods shows that GA, ANN and ANFIS can preserve optimal performances over wide range of disturbances with superiority to GA in terms of settling time using Integral of Square of Errors (ISE) criterion as fitness function.展开更多
文摘Presents the fuzzy neural network optimized by hybrid coded genetic algorithm of decimal encoding and binary encoding, the searching ability and stability of genetic algorithms enhanced by using binary encoding during the crossover operation and decimal encoding during the mutation operation, and the way of accepting new individuals by probability adopted, by which a new individual is accepted and its parent is discarded when its fitness is higher than that of its parent, and a new individual is accepted by probability when its fitness is lower than that of its parent. And concludes with calculations made with an example that these improvements enhance the speed of genetic algorithms to optimize the fuzzy neural network controller.
文摘Aiming at the problems that fuzzy neural network controller has heavy computation and lag,a T-S norm Fuzzy Neural Network Control based on hybrid learning algorithm was proposed.Immune genetic algorithm (IGA) was used to optimize the parameters of membership functions (MFs) off line,and the neural network was used to adjust the parameters of MFs on line to enhance the response of the controller.Moreover,the latter network was used to adjust the fuzzy rules automatically to reduce the computation of the neural network and improve the robustness and adaptability of the controller,so that the controller can work well ever when the underwater vehicle works in hostile ocean environment.Finally,experiments were carried on " XX" mini autonomous underwater vehicle (min-AUV) in tank.The results showed that this controller has great improvement in response and overshoot,compared with the traditional controllers.
文摘航空蓄电池作为飞机应急电源与启动系统的核心部件,其荷电状态(State of Charge, SOC)的精准估算对飞行安全与能源管理至关重要。针对传统单一算法(如安时积分法、扩展卡尔曼滤波(EKF)、神经网络法)在复杂工况下存在的模型依赖性强、非线性拟合不足或累积误差大等问题,文章提出一种融合卡尔曼滤波与神经网络的复合算法。该算法以二阶RC等效电路模型为基础,通过卡尔曼滤波实现状态递推,并利用长短期记忆网络(LSTM)补偿模型参数漂移与非线性动态误差。实验基于某型航空镍镉电池实测数据(涵盖充放电循环、温度波动及老化场景)。结果表明,复合算法的SOC估算均方根误差(RMSE)降低至1.2%以内,较单一EKF与LSTM算法分别提升42%与35%,有效解决了航空场景下高动态、宽温域工况的SOC估算难题。
文摘A novel approximation algorithm was proposed for the problem of finding the minimum total cost of all routes in Capacity Vehicle Routing Problem (CVRP). CVRP can be partitioned into three parts: the selection of vehicles among the available vehicles, the initial routing of the selected fleet and the routing optimization. Fuzzy C-means (FCM) can group the customers with close Euclidean distance into the same vehicle according to the principle of similar feature partition. Transiently chaotic neural network (TCNN) combines local search and global search, possessing high search efficiency. It will solve the routes to near optimality. A simple tabu search (TS) procedure can improve the routes to more optimality. The computations on benchmark problems and comparisons with other results in literatures show that the proposed algorithm is a viable and effective approach for CVRP.
文摘This paper presents an proportional integral (PI) based voltage-reactive power control for wind diesel based decentralized hybrid power system with wide range of disturbances to demonstrate the compensation effect on system with intelligent tuning methods such as genetic algorithm (GA), artificial neural network (ANN) and adaptive neuro fuzzy inference system (ANFIS). The effect of probabilistic load and/or input power pattern is introduced which is incorporated in MATLAB simulink model developed for the study of decentralized hybrid power system. Results show how tuning method becomes important with high percentage of probabilistic pattern in system. Testing of all tuning methods shows that GA, ANN and ANFIS can preserve optimal performances over wide range of disturbances with superiority to GA in terms of settling time using Integral of Square of Errors (ISE) criterion as fitness function.
